نشریه اکتشاف و پردازش هوشمند دانش
پیاپی 8 (بهار 1402)

A self-adaptive system can modify its own structure and behavior at runtime based on its perception of the environment, of itself and of its requirements. One key element of a self-adaptive system is its self-adaptation logic that encodes when and how the system should adapt itself. When developing the adaptation logic, developers face the challenge of design time uncertainty. To define when the system should adapt, they have to anticipate all potential environment states. However, anticipating all potential environment changes is infeasible in most cases due to incomplete information at design time.Online reinforcement learning (RL) addresses design time uncertainty by learning the effectiveness of adaptation actions through interactions with the system’s environment at run time, thereby automating the development of self-adaptation logic. Online-RL for self-adaptive systems integrates the elements of RL into the MAPE-K loopExisting online RL approaches for self-adaptive systems represent learned knowledge as a value function, so exhibit two shortcomings that limit the degree of automation: they require manually fine-tuning the exploration rate and may require manually quantizing environment states to foster scalability. In this paper, use policy-based deep reinforcement learning, which are structurally quite different, to automate the aforementioned manual activities. Deep RL addresses these disadvantages by representing the learned knowledge as a neural network. learned knowledge is hidden in the neural network. The results of the experiments indicate a high convergence speed of learning.

The most common types of cardiovascular diseases include congenital heart disease, heart failure, cardiomyopathy, rheumatic heart disease, pulmonary stenosis, and coronary artery disease. The diagnosis of cardiovascular diseases through symptoms is a major challenge in the current global conditions, and if not diagnosed in a timely manner, it can be a cause of death. Due to limited access of cardiac specialists to remote areas, an intelligent decision support system with fog computing approach can be an effective solution to improve the diagnosis of cardiovascular diseases. The aim of this article is to present an intelligent decision support system to enhance the diagnosis of cardiovascular diseases based on fog computing. Initially, medical information and records of 100 patients are collected. Then, the input data need to be cleaned and normalized, and features are extracted, selected, and weighted using feature extraction algorithms. Subsequently, the data is classified using a support vector machine algorithm. The results demonstrate that the accuracy metric in the proposed methods is higher than the baseline article. The proposed memetic-genetic method achieves an accuracy of 93%, while the graph mining-genetic method achieves an accuracy of 91%, compared to the baseline article with an accuracy of 86%. The proposed methods have been able to achieve a higher level of coverage in the ROC curve compared to the baseline article. In the memetic-genetic method, the TPR value is 0.38 at the zero point, while in the graph mining-genetic method, the TPR value is 0.2 at the zero point. In contrast, the baseline article considers a TPR value of zero at the zero point. Furthermore, the memetic-genetic method can reach a TPR value of one faster. The point where it reaches a value of one is 0.4. The proposed methods outperform the baseline article in terms of mean square error, and for the combined memetic-genetic and graph mining-genetic method, they achieve error values of 11% and 15%, respectively.

E-learning tools should be developed based on learners' psychology. Many studies have been conducted on the nature of learning and the factors affecting it. One of the psychological issues that should be considered is the user interface in E-learning because the user interface (UI) is the point of interaction between the user and the learner. The user interface influences the possibility of using an application and the ease of learning. If such correlation fails, the training objectives may not be achieved, even if the training content is well chosen and the user is willing to learn. For this reason, the main issues in successful collaboration, which should be considered in designing user interfaces for E-learning and educational applications, are the focus of this study. This research aims to investigate the rules of graphic user interface design for the design of Iranian educational applications. The research method is a descriptive survey, and the data collection methods are library-based and qualitative. One hundred educational applications published in the Google Store, rated between 4 and 5, have been examined in this study. These applications have been reviewed regarding overall design style, fonts, primary colors, icons, and more. Also, a questionnaire was designed to know the importance of each item from the audience's point of view, and it was distributed among 169 people, and its results were analyzed and checked using SPSS. The results showed that successful applications use similar design styles, fonts, colors, and icons, which can be used as a suitable model.

The current research examines the relationship between organizational culture and the quality of work life and customer relationship management among social security organization employees, which is new in its kind. In this research, Robbins' model was used to measure organizational culture, Walton's model was used for work quality, and Sin et al.'s model was used for customer relationship management (CRM). The statistical population of the research includes the employees of Khorasan Razavi Social Security Organization in the seven branches of Mashhad, whose number is 750 people, and 254 people are considered for the sample size of the research according to Morgan's table. The tool used to compile the questionnaire was a standard one, which has the necessary validity and its reliability was calculated using Cronbach's alpha coefficient, which shows the internal consistency of the questionnaires. The results obtained based on the structural equation model test indicated the existence of a significant and positive relationship between organizational culture and customer relationship management and the quality of work life, as well as a significant and positive relationship between the quality of work life and customer relationship management. In addition, the results showed that the quality of working life mediates the relationship between the subjective well-being of shareholders and the management of relationships with customers.

One of the most important priorities in developed countries is the use of machine decision-making instead of a human. One of the areas that need this field is health. For this purpose, determining the obesity and thinness of people can be very useful in studying and examining the health status of a society and adopting health system policies. Images of people as a database of research have been prepared from several different environments where the distance between the camera and the person is the same in all of them. Then, the background of the image is removed using background subtraction. Image features that include image morphological characteristics are extracted from the image and are classified into two categories to perform classification operations. The people were divided into three categories: fat, medium, and thin. The images are noised using the Gaussian low pass filter method with different frequencies filtered using two methods of salt and pepper noise and Gaussian noise. n normal images, the highest accuracy is related to the support vector machine method with an accuracy of 91.7%The results of this paper showed that with the proposed method, in addition to being able to classify the people of a society in terms of obesity and thinness, a higher accuracy was achieved than most of the methods that have been presented so far. According to the solutions and results of this research, by increasing the images of people, in addition to increasing the accuracy, it will reach a more practical level.

Today, heart disease is one of the main causes of morbidity and mortality. Since the initial and early diagnosis of this disease is very important and vital and the usual methods used in the medical industry need to spend a lot of time and money to diagnose this disease, accurate prediction of this disease has become a challenge. According to the huge amount of hospital data, which is added to its volume every day, the importance of data mining, which is one of the important techniques for discovering knowledge and hidden patterns, is increasing. Many studies have been done based on data mining to predict heart disease; according to their solution, each one pursues goals such as increasing speed, increasing accuracy, reducing the volume of calculations, and error coefficient. This research aims to increase the reliability and accuracy of heart disease diagnosis using the feature selection technique by meta-heuristic algorithms to extract useful features and to reduce the computational burden, and we use machine-learning algorithms to evaluate the proposed method. The results indicate that the proposed system can diagnose individuals with cardiovascular disease with relatively high levels of accuracy and precision. By improving the efficiency and accuracy of heart disease diagnosis, this research may contribute to better patient outcomes and reduced healthcare costs.